Slope comparison method (SCM) for the determination of trace amounts of silicate in ultrapurified water [An article from: Analytica Chimica Acta]

This digital document is a journal article from Analytica Chimica Acta, published by Elsevier in 2005. The article is delivered in HTML format and is available in your Amazon.com Media Library immediately after purchase. You can view it with any web browser.

Description:
A sensitive analytical method for the determination of trace amounts of silicate in ultrapurified water was developed. The method is based on the formation of an ion associate of molybdosilicate with malachite green (MG) and the collection of the ion associate on a tiny membrane filter (diameter: 5mm, and effective filtering diameter: 1mm). The ion associate formed on the membrane filter is dissolved together with the membrane filter in 1ml of methyl cellosolve (MC) and the absorbance of MC solution is measured at 627nm by a flow injection-spectrophotometric detection technique. In this method, silicate in the original sample (ultrapurified water) is concentrated as the ion associate into a small volume of MC to get high sensitivity. As sample concentration takes place, the small amounts of silicate contained in the reagents used also become concentrated as the ion associate into MC. The original sample volumes are varied and evaporated to an identical volume. Therefore, the reagent added is fixed to the same volume. The absorbance increase linearly with increase in the original sample volume will be due only to silicate in the original samples (ultrapurified water). The resulting slopes obtained by varying the sample volumes are compared with the slope of the calibration graph, and thus named the slope comparison method (SCM). The SCM facilitates a more sensitive and accurate evaluation of silicate concentration in the samples than either common calibration method (CCM) or standard addition method (SAM) because it compensates for the influence of trace amounts of silicate contained in chemicals, reagent solution and solvent used. The calibration graph was constructed from 0 to 0.25ngml^-^1 of Si and the detection limit was 10pgml^-^1 (ppt) when 30ml of samples was used. The standard deviation and relative standard deviation from six measurements of the reagent blanks were 0.0012 and 3.5%, respectively.